Study finds pharmacist involvement improves quality

But cost savings are elusive

When hospital pharmacists push for increased pharmacist involvement in clinical care, the goal often is to improve the quality of care and cut rehospitalizations and related costs.

One new study has added to the body of evidence that clinical pharmacists play an important role in improving the quality of patient care, but its findings stop there: The research found no evidence of improved costs or readmission rates.1

"The literature cites as many as 60% of problems that occur post-discharge could be medication-related, and 20%-30% of discharged patients have problems 3-5 weeks post-discharge," says Paul C. Walker, PharmD, FASHP, clinical associate professor in the department of clinical, social, and administrative sciences at the University of Michigan College of Pharmacy and the assistant director of clinical services in the University of Michigan Health-System in Ann Arbor, MI.

The new study did find that having a pharmacist involved in medication reconciliation at discharge improves the quality of patient discharge by identifying and resolving medication discrepancies.1

"Since discharge plays such a big role in patient's outcomes, we thought if we could identify problems at the point of discharge and resolve them then patients could go home, and we'd prevent readmission," Walker explains.

To Walker's and co-investigators' surprise, this did not occur.

The study found that medication discrepancies at discharge were identified in 33.5% of intervention patients and 59.6% of control patients. All discrepancies were resolved in the intervention group prior to discharge, but study results showed no significant difference in readmission rates between the intervention group and control group at 14 days and 30 days. Also, there was no difference in emergency department visits.1

"I think one of the reasons we didn't see an impact on those economic outcomes is because our patients are pretty sick, and there are likely to be other variables that impact hospital readmission that we could not control for in our study," Walker says.

The project was developed from a grant the University of Michigan received for a demonstration project from the Centers for Medicare and Medicaid Services (CMS).

"It was to look at how health care for Medicare recipients could be improved across transitions in care," Walker says. "We got involved to find out what the pharmacy could do to help with the process."

The hospital already had plans to improve on its discharge process before the CMS demonstration project came about, notes Christopher Kim, MD, MBA, FHM, assistant professor of internal medicine, pediatrics, and assistant medical director for the faculty group practice, and assistant chief of staff for the office of clinical affairs at the University of Michigan Health System in Ann Arbor, MI.

"What was serendipitously happening is the pharmacy department had ready-made plans to geographically place clinical pharmacists closer to the unit," Kim explains. "So now clinical pharmacists are closer to the units where patient care is provided, and they're available to not only provide answers to questions from the nurse, but also available to go in and speak directly to the floor staff and patients."

After a brainstorm session, hospital pharmacists decided to have a pharmacist involved with medication reconciliation at discharge, he adds.

"We thought by having a hospital pharmacist provide medication reconciliation and give patients information about the medications they will need at home, plus do a follow-up by phone after discharge, they might reduce the 30-day readmission rate or reduce the use of emergency services," Walker explains.

"The whole issue of transitional care is a hot topic right now," he adds. "With Medicare across the country, we're seeing rates of 20%-30%, so that was the emphasis behind our work."

Having a pharmacist involved with medication reconciliation and follow-up is expensive.

"The average salary of a pharmacist is $90,000-$100,000 in our area, plus benefits," Walker says. "So we're talking about $135,000 a year to have the pharmacist involved in the process."

The study could not demonstrate a cost offset because the pharmacist's involvement was not shown to reduce readmission rates or ED visits, he adds.

"We funded this project out of existing hospital dollars, so support was not continued when the study was over," Walker says.

On the other hand, the pharmacist's involvement was well-received by patients, he notes.

"Patients were very satisfied and appreciated pharmacist interaction," Walker says. "We did a customer service survey to get feedback, and we found they were very appreciative of the service."

The intervention had a pharmacist identify high-risk patients by looking at the number of prescribed medications and the number of changes made following admission.

"We looked at patients who had the handful of diagnoses that the CMS project was looking at, including diabetes and congestive heart failure," Walker says. "It turned out that based on our risk factors we had a large number of patients who were eligible for enrollment in the study."

There were three targeted medical services, and the pharmacist would work with one or two services for a month and then move to other, he explains.

The pharmacist did this in alternating fashion, so patients who were not receiving the pharmacist's services that month were receiving usual or customary care, Walker says.

The pharmacist also worked daily with the attending physician and the entire medical team, including the discharge planner.

"She'd review the medical record and reconcile medications with the patient, identifying any potential discrepancies," Walker says. "She made sure prescriptions written at discharge were the right ones, and she'd provide patient education to the patient and caregiver, going over the medication list."

The pharmacist used the teach-back method to see if patients understood the information.

If the patient was prescribed a new medication, the pharmacist made certain the patient understood how to use it, and, in some cases, facilitated getting the prescriptions filled.

Then after the patient returned home, the pharmacist would call within 72 hours to see if there were any problems related to medication or any concerns about the discharge, Walker says.

"She'd work to resolve any issues," he adds. "If the problem was not medication-related, or if it was an issue the pharmacist couldn't resolve, then she'd triage the patient to the prescribing physician or nursing staff."

The idea was to identify any problems and refer the patient to the appropriate medical provider to receive assistance.

In one unpublished finding, investigators found in conversations with patients post-discharge that some still had medication-related issues despite the pharmacist intervention, he says.

"If we gave them instructions they might not recall them," Walker says. "That raises the question about when is the best time to teach patients about therapy; is it at discharge, or earlier in their hospitalization, or after they go home?"

One problem might be that hospitals barrage patients with a lot of information at discharge, and they can't assimilate it all, he suggests.

"That would make a very interesting investigation," Walker adds.

The hospital now has nurses provide medication use instructions at discharge, but Walker and colleagues are working on a model that would also engage pharmacists in the education.

"It's being done in an interdisciplinary fashion, and the work is very new," he says. "We're looking at other ways to improve the transition of care to improve readmission rates, and one is to have each health care provider give his or her own expertise to the discharge process."

Although the hospital has stopped funding a pharmacist position for medication reconciliation, other process improvements have been implemented, including an electronic process for medication reconciliation at discharge, Walker says.

"When physicians are discharging patients who are in our system, they can look at the electronic medication list, both for inpatient as well as for the outpatient world," Walker says. "They see both lists and can update the lists and generate any necessary prescriptions."

Investigators are evaluating that system to see how well it is working, he adds.

"I have another project going with a colleague to look at whether it makes a difference to have a pharmacist make phone calls after the patient goes home to identify medication-related issues," Walker says.

Investigators are continuing to look for ways pharmacists can contribute to an improved hospital discharge process.

There might be some more cost-effective models, involving hospital pharmacy personnel, he notes.

"In some cases pharmacy technicians can do this work," he says. "This would reduce the volume of work pharmacists have to do, and it could improve the transition of care and hopefully reduce our readmission rate."


  1. Walker PC, Bernstein SJ, Jones JNT, et al. Impact of a pharmacist-facilitated hospital discharge program. Arch Intern Med2009;169:2003-2010.